About 1 in 3 people newly prescribed DPP-4 inhibitors will be non-adherent or stop taking the drug altogether within a year, Australian figures show.

A review of PBS data for almost 16,000 people with type 2 diabetes found that 36% were non-adherent (taking less than 80% of the prescribed doses) and 30% were non-persistent (discontinued for at least 90 days) at 12 months.

Switching between different DPP-4 inhibitors was uncommon, with only 3% of patients overall changing from one agent to another within a year, according to researchers from Monash University, Melbourne.

Their analysis of PBS dispensing data for the period of January 2015 to August 2017 covered new prescriptions for adults taking sitagliptin (n= 9576),  vildagliptin (1130), saxagliptin (1126), linagliptin (3560) and alogliptin (523).

With the exception of saxagliptin, there were no differences in non-adherence and non persistence rates between the different DPP-4 inhibitors.

Compared with the index drug sitagliptin, saxagliptin was associated with a higher likelihood of being non-adherent (43.4%; odds ratio 1.41, 95% CI 1.23-1.60) or non-persistent (35.9%; hazard ratio 1.27, 95% CI 1.15-1.42).

The study authors said that in randomised controlled trials the gliptins had shown better adherence rates than other drug classes such as sulphonlyureas, thiazolidiones and GLP-1 inhibitors.

But the real world adherence rates seen in the current study suggested that many patients prescribed DPP-4 inhibitors would not achieve the reductions in HbA1c levels expected from trials, they said.

“Interventions aimed at addressing the modifiable risk factors of non-adherence and non-persistence (such as health literacy, quality of the relationship between patient and physician) to DPP4-inhibitors are required,” they wrote.

And although the overall rate of switching within class was low (3.2%), in the multivariable Cox regression model, vildagliptin (HR 1.60), saxagliptin (HR 2.44), linagliptin (HR 1.38) and alogliptin (HR 2.42) were associated with higher likelihoods of switching compared to sitagliptin (2.3%).

The study authors said the low rate of in-class switching may be due to the fact that there are few differences in clinical effectiveness between individual DPP-4 inhibitors, particularly in terms of achieving change in HbA1c levels or body weight, or their risk of hypoglycaemia, and adverse drug events such as acute pancreatitis or cardiovascular risk.

“Thus, overall, clinicians and patients may perceive low clinical utility in switching from one DPP-4 inhibitor to another.”

The study is published in Diabetes Research and Clinical Practice. It was not funded by any specific grant from public  or commercial funding sources. The study co-authors declared interests in advisory boards and/or receiving honoraria from several pharmaceutical companies for work unrelated to the study.